RCM 4

Download as pdf or txt
Download as pdf or txt
You are on page 1of 32

Implementation of Reliability Centered Maintenance

(RCM) in the background of industry 4.0 –Issues,


Challenges, and Opportunities
Rahman Khalil Ur
Northwest Normal University
Zhibao Mian (  Z.mian2@hull.ac.uk )
University of Hull
Ali Yasir
KPK

Research Article

Keywords: Industry 4.0, Reliability-Centered Maintenance (RCM), Intelligent Maintenance

Posted Date: June 21st, 2023

DOI: https://doi.org/10.21203/rs.3.rs-3053231/v1

License:   This work is licensed under a Creative Commons Attribution 4.0 International License.
Read Full License

Page 1/32
Abstract
Reliability Centered Maintenance (RCM) is a well-known maintenance method originated in the aviation
sector. It has shown a huge success such that consequently this technique has been adapted in more
industrial sectors. Currently, the system maintenance has become more extensive and complex task. As a
result, the equipment structures and functions are continuously becoming more complicated and diverse,
which ultimately makes determining their reliability much more challenging. In addition, traditional RCM
techniques can no longer meet the new requirements, including real-time condition monitoring, health
diagnosis, remote maintenance, and intelligent maintenance required in industry 4.0. This paper
highlights the challenges, issues and difficulties in adopting RCM with industry 4.0. A conceptual model-
based solution is put forward to provide solutions towards the identified challenges. The model-based
solution includes effective and promising solutions by combining RCM technology with artificial
intelligence, digital twins, internet of things, big data, and model-based reliability analysis technology to
provide advanced real-time data based equipment management, including online health monitoring and
intelligence fault diagnostics and remote maintenance. This study will provide a base for implementing
RCM with Industry 4.0 to provide smart and intelligent maintenance of the equipment’s or assets.

1. INTRODUCTION
Maintenance has become a major challenge for the companies over the last few years. Therefore, the
modern companies are seeking ways to uplift the maintenance at advanced level by applying different
strategies and techniques. Among the strategies, Reliability Centered Maintenance (RCM) has been
around for decades to provide maintenance to the various organizations. The concept of RCM, initially
originated in the airline industry in the 1960s [1]. It is a standard procedure for determining the most
appropriate maintenance type and frequency for specific equipment units. RCM is a technique of finding
the most effective maintenance methods [2]. It can be proven more effective in maintenance planning
which the most essential pillar of sustainability [3]. Similarly, it has the ability to provide a monitoring
mechanism to concentrate on safety and business due to its characteristics of reliability [2]. It has shown
good results, such as in Indonesia, RCM has been adopted to improve the machines maintenance,
ultimately leads to reliability and productivity in automotive industry [1]. Another application of RCM has
been witnessed in electrical generation company to keep the assets working within premises of designed
working by reducing the probability of failures [5]. RCM procedure can also be applied to find out which
maintenance action is more optimal in reducing the cost, availability and safety [6]. RCM can be applied
to provide systematic analysis [7]. RCM has several applications such as distribution system,
transmission and power generation [8, 9]. It has potential of leveraging energy area, where the service
continuity indicates mandatory requirements; and maintenance has to be optimized with respect to time
and money [10]. The major working core of RCM is known as Failure Modes and Effect Analysis (FMEA),
which helps in identifying and analyzing the possible failure modes related to components of system. It
assists in identification of most critical parts of machinery and possible failure mechanism, ultimately
leading to the parameters of system that to be monitored. FMEA works on the principal of dividing the

Page 2/32
machinery into functional systems and then analyzing the failure mode of each machinery based on
manufacturing functions. RCM is also integral part of Condition-Based Maintenance (CBM), where its
working mechanism to identify critical components, possible failures of components [11].

Besides, bringing many profitable benefits still there are some limitations and challenges in RCM’s
applicability that are required to be addressed before making it as maintenance solution. Similarly, the
working procedure of FMEA can be time consuming and it requires a sound knowledge about
understanding the nature of risk, prioritizing risk and its occurrence. Although, RCM techniques have
shown major focus on the optimization and cost analysis, however, still some research gaps exist and
more work is required to build an effective and sophisticated condition monitoring system [2]. Similarly,
maintenance efficiency has become a major challenge for the companies over the last few years due to
lack of progress such that many companies face a significant challenge of facing pressure related to
cutting the maintenance overhead [3]. Therefore, companies are hunting for new strategies or solutions to
adopt incremental or continuous maintenance approaches. The best potential solution can be the
application of tools and methodologies furnished by Industry 4.0 for maintenance process.

Keeping eye on the challenges faced by RCM during the maintenance, Industry 4.0 can be proven as
game-changer. Industry 4.0 is innovate concept of applying the cutting-edge technologies to provide
digital solutions through automation and digital exchange [4]. Industry 4.0 is interplay of many
technologies such as cyber-physical systems, Industrial Internet of Things, Artificial intelligence, Cloud,
advanced computers, Blockchain, Big data and Analytics, etc. This integration of various technologies
allows the seamless flow of data among the system components, increasing speed and reliability of the
operating machineries. It provides solutions to maintenance problems by shifting the maintenance from
preventive to predictive.

Similarly, the traditional RCM techniques or alone RCM might not be adequate enough to cope with
challenges and requirements related to maintenance such as real-time condition monitoring, health
diagnosis, remote maintenance, and intelligent maintenance required in industry 4.0. Thus, RCM can also
be proven effective, if it is combined with industry 4.0 to provide more comprehensive and intelligent
solutions towards the maintenance problems. However, still there are some serious challenges in the
blend of RCM and Industry 4.0 that are required to be addressed before making it an ideal platform for
maintenance. In this review, our prime focus to highlight the issues and challenges and final aim is to
present a model-based solution which can be helpful by providing solutions and opportunities for the
implementation of RCM in the background of industry 4.0.

1.1 Problem statement and research objectives


The integration of RCM with advanced technologies operating in industry 4.0 can be productive; however,
it still brings challenges that need to be addressed to take full advantage of its applications in the
background of industry 4.0 set up. Embedding RCM technology with industry 4.0 is a difficult task due to
major challenges of designing an interface, where model-driven engineering techniques including model

Page 3/32
transformation techniques such as RCM data has to be transformed into industry 4.0 e-maintenance
system; and nature of technologies such as IoT, Big data, Digital Twins, Blockchain and artificial
intelligence in Industry 4.0. For example, the challenges which may arise due to combining IoT and RCM
are related to privacy and information handling, designing a scalable architecture, heterogeneous source
of data, insecure software configuration and vulnerabilities etc. Similarly, integrating RCM with Digital
Twins may also create challenges such as resistance in technology adoption, security, mutual
understanding, interfaces; standardization and efficient information flow. Similarly, challenges created by
combing RCM with Big data include bulky data of maintenance, data acquisition, lack of expertise etc.

The major research objective of the proposed study to pinpoint the various challenges that can be
emerged due to integrating RCM with cutting edge technologies such as model-based reliability analysis
technology, digital twins, internet of things, and model-based reliability analysis artificial intelligence, big
data, technology in the industrial setup 4.0. RCM technology standalone cannot be effective in providing
reasonable solutions towards predictive and intelligent maintenance, health condition monitoring of
assets and health diagnosis of equipment or components in industry 4.0. Therefore, it should be
upgraded or modified before it to interplay with new emerging technologies such as IoT, DT, AI, etc.
covering all the issues in its deployment. This study also provides tentative solutions towards the
challenges by presenting conceptual model.

2. Related work
In this section, the major focus is to highlight the role of RCM for smart predictive maintenance. It also
covers the various models, techniques, methods, algorithms and applications using the current digital
technologies such as IoT, Big data, Digital Twins, Blockchain and artificial intelligence, presented for
different types of maintenance approaches in industry 4.0 environments.

Das et al. [5] Presented a conceptual model using RCM techniques such condition monitoring for
household appliances by using IoT infrastructure. The proposed model provides zero-failure
maintenances by integrating RCM and IoT.

Chris et al.[6] also presented IoT-based predictive maintenance (PdM) for rolling stocks. They exploited
RCM for the failure management. They applied RAMS (Reliability, Availability, Maintainability, Safety)
method for IoT-based condition monitoring.

Venkataswamy et al. [7] put forward study that is the combination of RCM, IoT and Failure Modes and
Effect Analysis (FMEA) which is intended for the distribution of electricity transformers. They developed a
trained model, where RCM receives FMEA and IoT data to predict RCM schedule for future instant. This
model is produced optimized maintenance schedule by using machine learning technologies.

Rahman et al.[8] provided a study that discussed the two well-known maintenance techniques such as
preventive and corrective within RCM. This paper is focusing on discussing the Society 5.0 and Industry
4.0 through the lens of Condition based monitoring and Machine learning and Artificial intelligence
Page 4/32
(MLAI) technologies.Jafarpisheh et al.[9] conducted a case study to present a decision making model of
reliability centered-maintenance for mining transportation machines in Esfahan Iran. Their work is
supported by using multi-criteria decision making method such as preference ranking organization
method for enrichment of evaluations (PROMETHEE) approach for assessment.

Compare et al.[10] outlined the major challenges related to IoT-enabled predictive maintenance in industry
4.0. They focused on building cost-effective, convenient and productive predictive maintenance
model.Achouch et al.[11] reviewed various tools and methods of intelligent predictive maintenance in
industry 4.0. They highlighted four major challenges in predictive maintenance such as organizational
and financial limits, data source limits, machine repair activity constraints and problems during the
deployment of industrial PdM models.

Hussain et al.[12] also introduced simulation model based on RCM approach for optimizing the
performance of CNC machine. The proposed method is applying RCM to find the reliability of machine
critical components for improving the productivity through planned preventive maintenance. FMEA is
utilized for the prioritization of failure of critical components. The major objective of this study is to
reduce the PPM time of CNC machine by using preventive maintenance.

According to our literature study not only IoT has been adopted along with the support of RCM for
predictive maintenance; but we also found a study that is related to utilizing Digital Twins (DT) in
integration with RCM in industry 4.0. In this regard, the study carried out by Mubarak et al.[13] introduced
the idea of implementing DT with RCM for predictive maintenance. The major consideration of the
proposed framework is cost and reliability in the industry 4.0 setup. The proposed optimized predictive
maintenance framework achieves accurate results for failure diagnostics and prognostics due to the
application of machine learning and advanced data analytics technologies for data fusion of PdM. The
role of data fusion of PdM is of significant importance. DT with integration of Industrial IoT have been
applied by Liu et al.[14] for the predictive maintenance of aircrafts. In this study, the DT has been applied
for modelling and simulation and IoT is adopted for data acquisition and processing for maintenance,
repairing and overhauling of aircraft maintenance.

We also highlighted some of related works that are utilizing artificial intelligence (AI) and machine
learning approaches for different kinds of maintenance tasks. Similarly, machine learning and artificial
intelligence has been used in collaboration with RCM in industry 4.0 for the purpose of making
convenient the maintenance task. In this regard, Cheng et al. [15] developed an intelligent RCM based
system that is leveraging artificial intelligence for the predictive maintenance. They introduced case-
based reasoning and rule-based reasoning methodologies for building intelligent RCM known as IRCMAS.
This intelligent model helps in reducing the price and improving the decision making for RCM analysts.
The integration of AI technologies also helps in minimizing the repeated tasks of RCM in the proposed
system. Similarly, Cheng et al.[16] also came up with idea of bringing AI technology in RCM process. They
presented intelligent RCM approach known as “intelligent RCM analysis system” (IRCMAS) to improve the

Page 5/32
efficiency, accuracy, and quality and validity conventional RCM process by leveraging case-based
reasoning (CBR).

The huge volume of historical data has always been the major challenge for the maintenance-based
approaches. To deal with this challenge, Li et al.[17] developed a machine learning approaches for the
predictive maintenance and operation management in railway system. The proposed model uses
machine learning technology to use the real-time and historical data to predict the future event, railway
network velocity and service interruption, conditions causing failure in future, cause of failure and wear
rate of component. In the literature study, we also investigated the complementary role Big data and RCM
for the maintenance jobs to address the challenges related to the volume of data. According to literature
study, we found that Saidi et al.[18] presented study, which is emphasizing the contribution of Big Data
for RCM. This research study highlights the three major challenge faced by RCM with respect to data
such as volume of data, source of data and frequency of processing or analysis of data. The role of Big
data with respect to dealing with already mentioned challenges, has been discussed in comprehensive
fashion to deal the diverse data collected from various sensors and actuators for the maintenance.

3 INTEGRATING RCM WITH INDUSTRY 4.0 AND CHALLENGES


The major purpose of this section is to highlight the role of RCM to provide massive support to the
maintenance in industry 4.0; and then to highlight the various challenges faced by RCM during its
integration with industry 4.0. The role of RCM along with different challenges have been discussed.

The various emerging technologies in the industry 4.0 present significant opportunities and challenges
for the application of RCM. Due to the widespread deployment of sensing systems and modern digital
technologies such as Big-Data, IoT and Artificial Intelligence, the production of huge enterprises is rapidly
changing in the recent years. Predictive Maintenance is becoming more important for the manufacturing
and production sectors because to the Industry 4.0 concept. PdM heavily relies on the Internet of Things
(IoT), which digitalizes physical behaviors by enabling connections between machines, people, and other
machines for intelligent observation. In the context of Industry 4.0, PdM employs sophisticated online
analysis of the data gathered for the early identification of probable equipment problems and assists
engineers with maintenance operations by providing guided smart decision assistance. Cyber-physical
systems (CPS), the Internet of Things (IoT), mobile devices, cloud computing, big data, and sophisticated
AI approaches are among the key technologies of Industry 4.0 [19, 20]. Cyber-physical systems and
sensors are also being used for digitizing physical processes and integrating control operations and IT-
based data-driven automation as part of Industry 4.0. Flexible automation, digital twins, predictive
maintenance and other supply chain management optimizations are just a few of the industrial uses
enabled by Industry 4.0. Industry 4.0 collects enormous amounts of digital data regarding the condition
of machinery and equipment in corporate maintenance. Sensors such as vibration, sound, temperature,
power consumption, and thermal cameras are used to collect this data. After collecting this huge amount
of data, it is required to assess the condition and maintenance of components operating in Industry 4.0.
This process can be achieved by using a smart and intelligent maintenance approach. Intelligence
Page 6/32
maintenance requires a deep understanding about the current condition of components and future
events, likely to be occurred, which is also known as predictive maintenance. Similarly, the reliability of
system is also a major consideration, while building an intelligent model in the industry 4.0. For this
purpose, the implementation of RCM can be proven affective as it take into account the reliability of
system. However, integrating RCM with Industry 4.0, some key challenges that need to be addressed.
Identifying and overcoming these challenges will necessitate the digitization of production processes and
the implementation of intelligent automated systems that can self-manage and produces the most
productive outcomes.

This study will mainly discuss issues and challenges through the techniques involved in industry 4.0,
including "Digital Twins in industries, Industrial Internet and 5G, Industrial Big Data, Industrial Internet and
Internet of Things, Industrial artificial intelligence, Cloud manufacturing and cloud edge collaboration,
Industrial Blockchain, eMaintenance Challenges, Model-based maintenance". The integration of RCM with
different technologies operating in Industry 4.0, along with future challenges are discussed as below.

3.1 IOT-RCM INTEGRATION AND CHALLENGES


IoT has changed the working and shape of industrial setup by providing more advanced automated
operations and predictive maintenance in industry 4.0 [8]. But, still maintenance can be challenge, where
the components can be monitored to keep them alive, productive and risks-free. Therefore, RCM can be
adopted in IIoT to provide smart maintenance by reducing risks and maximizing the productivity with the
support of digital transformation. In this regard, duo technologies i.e. IoT and RCM based model has been
presented by Venkataswamy et al.[7] for distribution of transformers. Similarly, IoT can be proven more
effective by providing smart options of predictive maintenance and analytics because IoT sensors have
been witnessed to be more viable as they collect more real-time data about event or conditions. Access to
the data and analytics can provide high level of RCM by eliminating risks and enhancing the productivity.
By implementing RCM with IoT in industry 4.0, it will also reduce the cost and availability will be
increased by keeping machinery safe from risks and augmenting productivity. In IoT-prescriptive
maintenance the RCM can be applied to enhance to analyze and measure the important information and
assets[21]. RCM, in prescriptive maintenance enables the maintenance manager to identify the parts and
components of critical assets that required to be monitored by leveraging IoT sensors, the types of failure
modes in the critical part or component, and the data and the various ranges to be collected. It can
accelerate and optimize the procedure of maintenance by capturing the most relevant data and
measuring the performance of the components and assets that are crucial with respect to manufacturer’s
operations.

Besides, these benefits there exists some challenges that are needed to be addressed as discussed
A. Limited applicability
IoT has been an important technology in enhancing the maintenance efficiency, however, the application
domain of IoT are limited to aircraft and building instruction fields. Similarly, the application of IoT has

Page 7/32
been mainly focused only on remote monitoring systems [6]. In addition to, in condition-based monitoring
a further development is required due to lack of high fidelity related to faults [5].

B. Security challenges
Similarly, security is also a major concern[22]. However, the privacy issues raised are similar to those
raised by "Big Data". IoT devices must be protected from external attacks, such as Malware and
intrusions. IoT devices are increasing at an explosive rate, but many challenges exist. Among the hurdles
the significant technical hurdle companies face with IIoT predictive maintenance systems is related to
when they run into data bottlenecks where the response is significantly is slowed down[23]. This can be
avoided by removing delay or friction during the generation of data as results of insights trigger action in
the field. This challenge can be more significant, when more measurement and sensors will send data to
wireless networks [10]. Data privacy is also a major threat, while integrating real-time data from multiple
data sources.
The information handling and privacy are still a big challenges related to IoT; and they will be even more
significant, if combined with blockchain. Insecure software or firmware is another source of IoT security
threats. New vulnerabilities will likely arise, if a company sells a product with the most recent software
update [24]. IoT devices are increasing at an explosive rate, but many challenges such as lack of anomaly
features detection in intrusion detection system[25], simple password and insecure interface [26], default
configuration, updating irregular software, a gap between patch releases and installation [27] etc. These
security vulnerabilities will drastically affect the overall reliability and performance of the industry set up.

C. Reference architecture and standardization


A higher level of abstraction description that helps to identify problems and difficulties for various
application scenarios is the IIoT reference architecture. The scalability, extensibility, flexibility, and
interoperability across heterogeneous equipment employing various technologies must be highlighted in
the IIoT design architecture. Numerous reference architecture frameworks were developed for this aim in
various IIoT application scenarios. [28]. IIoT is meant to handle vital machinery that can fail if control
signals are timed incorrectly by as little as a few thousand of a second, network performance
characteristics like latency—the time it takes for a message to traverse the network—are crucial. Shutting
down a machine in response to a safety concern detected by a sensor is useless if the shutdown
instruction arrives a few seconds too late. To fully realize the IIoT potential, global enterprises must
communicate with thousands of kilometers apart devices. To fully utilize, IoT based industry, several
standardization and consortia have been initiated at every layer of IoT.

D. Fault predication
In smart IoT based system, the fault and historical based prediction can be challenging job to provide the
accuracy and correctness of the industrial maintenance [29].

3.2 DIGITAL TWINS –RCM INTEGRATION AND


CHALLENGES

Page 8/32
According to our literature study, both technologies Digital Twins and RCM have been applied in the
predictive maintenance. Digital twin has significant role in the predictive maintenance [30–32]. Similarly,
RCM provides reliable and cost-effective predictive maintenance and optimized decision making [8, 13].
RCM can be integrated into digital twins by providing more intelligent and smart maintenance; and
reliability of assets. In the data-driven reliability, the collected and contextualized data can be provided to
the RCM for making better strategic decision and forecasting. Similarly, Digital Twins (DT) is the
emerging technology that empowers the asset managers to make better and prompt decisions about the
asset reliability. The various involved in digital Twins before making it combination with RCM that are
given below as.

A. Complex Modeling and Data flow


For machines functioning in complex working environments, using digital twin for predictive maintenance
in open system architecture is essential to avoiding unexpected and catastrophic machine breakdowns
and saving production downtime. To use a digital twin to address a predictive maintenance problem, the
problem must be predictive with relevant operating history, and have domain specialists who understand
the problem. Digital twin capabilities are provided by vendor-agnostic frameworks, cloud vendors, and
industrial manufacturers. The problem should include a record of the equipment's operational history,
including positive and negative outcomes. The recorded history should be relevant to the use case and of
sufficient quality to support it. Finally, the company should have domain experts well-versed in the issue.
Consequently, the current virtual models can be failed to accurately describe physical entities in objective
manner and realistic manner. A simulation is a valuable tool for solving the real world problems[33].
During the last few years, a significant attention has been paid due to the fact, it provides decision
making techniques in environment, where stochastic variables are involved [34]. Still, digital twins lack
simulation tools and mathematical models[35, 36]. The production of enormous operational data
presents new diagnostic and prognostic challenges. Algorithm creation, feature extraction, performance
optimization, and other aspects of Bigdata based diagnosis and prognosis will be the standard which is
another big issue.

B. Data analysis issues


Digital Twins introduces several challenges related to big data, such as data variety and issues related to
its dynamics [36]. The efficient processing, analysis and storing bulky volumes of data in a DT is still an
open challenge[37]. Therefore, before, creating any DT simulation, it is important that data should be
provided in high quality, cleaned and sorted structure. Highlighting such a challenges will indicate the
importance of researching an efficient data management solution that assures the encapsulation of DT
domain information, addresses the data structuring complications for D.T and allows users to develop
DT data models without having complete domain expertise [38].

C. Security challenges
Security challenges the less attention paid to security at the onset of a new type of technology, the faster
it spreads. Because digital twins are hosted in the cloud and don't require any physical infrastructure, the
security issues are smaller than with traditional systems. As a result, data encryption and intrusion
Page 9/32
detection systems will become critical components of DT. Other worries about data sharing remain,
including security, privacy, accountability, ethics, and responsibility. Agreement frameworks exist in the
geospatial domain, but they only work in government-dominated environments where data exchange is
encouraged [39]. The primary challenges and possible benefits of Industry 4.0 and digital twins for
seamless product integration, as well as manufacturing and specification methods that connect digital
twins to other domains, may offer up new alternatives for tolerable allocation [40].

D. Technology adoption, culture, and lack of skilled person


Although, digital twins can be a best possible option for simulation and representing the physical world in
digital pattern with reliability and knowledge-based accumulation model through the application of
reliability centered strategies for the maintenances in industry 4.0. But, the proposed model based on DT
must be possessing some specific requirements such as scalability, expansibility and fidelity in industry
4.0 setting [41]. Digital Twin can create more intelligent and fidelity-based models [42]. But, still there are
some barriers highlighted by the study presented by Broo et al. [43] during the creation of such models
which are leveraging DT are such as adoption of technology; culture; and lack of technical and skilled
people. To deeply understand the various challenges in incorporating DT technology, Broo et al. [43]
conducted a survey-based study, where participants recorded the previously mentioned three major
challenges. The first issue is related to the ignorance of people or to let people to change behavior or
requirements and understand the value of data. The second issue is about the cultural constraints such
as political barrier in implementing the technology. The last one is related to highlighting the importance
of creating business use-cases and lack of skillful personnel to run such activities. Some important
considerations or factors i.e. mutual understanding, interfaces; standardization and efficient information
flow should be taken into account before integrating DT with any other advanced technology for building
framework [40].
3.3 INDUSTRIAL BIG DATA-RCM AND CHALLENGES
Big data analytics is not only important in the current scenario but can be used for futuristic purposes as
well. Amazon is using Big Data analytics with help of predictive forecasting enable the pre-shipping of
products to the distribution centers near to final customers before the order can be placed [44]. Data
analytics with support of RCM can be applied to critically analyze an asset or component to get complete
insight about the maintenance and operation. Data has significant in RCM strategy, as it powers the
companies to integrate multi data sources into single seamless dashboard, which enables them to take
communicating actionable insights and make the right choice of technology/devices. This decision
making results in labor cost reduction, assessing the asset downtime and catastrophic failures supported
by enhanced efficiency.

The combination of RCM with Big data can be effective apparatus for solving maintenance issues. RCM
has the potential to provide reactive, proactive, preventive and predictive approaches of maintenance
about assets. In this regard, the role of data is of utmost importance particularly in case of proactive
approach. By incorporating a sophisticated data analysis mechanism can be assisting in building smart
decision making maintenance and understanding the current operations. Every machine or piece of
Page 10/32
equipment failure has the potential to interrupt the supply chain, therefore maintenance focuses on
decreasing and eliminating the amount of failures that happen during production. The maintenance
policy's purpose is to give guidance in selecting the most economical maintenance method and system
to guarantee operational safety. For important components, predictive maintenance is strongly advised
since their failure would pose a serious safety risk in addition to significant function loss. Big data and
related approaches have significantly increased the speed and accuracy of maintenance decision-
making, as well as the transparency of the system health [45]. Along with developing big data techniques,
various data, mainly asset operating, including failure data can be detected, collected, and analyzed for
fault diagnosis and failure prediction in RCM. The most first challenge is related the system definition
and system modelling for this type of system investigation. Another challenge in the big data processing
is related to the dynamic behavior of the system[37].

Another challenge of maintenance in the face of the volume of data generated of significant challenges
of maintenance is the explosion in the volume of data, for example, a variety of information (several
sources unstructured and structured with different formats) [36]. One of the significant challenges for
PdM is the limited number of professionals with expertise in the fields of Maintenance Data System
(MDS), data science, specific technical skills; and the hurdles to train the predictive maintenance model
without disclosing business data [46, 47]. Undoubtedly, the role of data is of prime importance in the
current time and future as the type and amount of data in the industry is significantly growing. Design
data, such as data from the product and machine design, Machine operation data, such as data from the
control system, equipment operation, Environmental conditions, fault detection, weather information,
indoor temperature, humidity, noises and data about system status monitoring [48].

The manufacturing and production logistic databases can manage large amounts of data using the big
data platform. It results in the creation of Computerized Maintenance Management Systems (CMMS),
which facilitates the formulation of maintenance strategy decisions. PdM has been effectively
established by studying the machine degradation process and symptoms spanning from typical to
defective scenarios in order to propose a corrective solution for the random machine failure in
maintenance management [58]. The PdM module has two key systems, such as a prognostics system
and a diagnostics system, for integrating the sensor knowledge with the real machine. One effective
method for determining machine/component failure is PdM analysis. Additionally, it provides a startlingly
precise future breakdown utilizing time-series data and sophisticated analytical techniques. The PdM
module carefully supervises the relentlessly aids the analytics process and machine condition is related
to diagnosis and prognosis [45].

Manufacturing industries use sensors and data capture systems to run their physical facilities, generating
a large amount of data to process, referred to as Big Data [49]. The capacity to detect patterns in data
that are indicators of upcoming defects requires the availability of historical sensors' real data (for both
faulty and healthy system situations). Due to the huge data sets, data size variables constantly change,
ranging from terabytes to petabytes to Exabytes or Zettabytes [50]. Big Data techniques in the scope of
PdM comprise, in particular, how to store and process vast and complex data obtained from sensors, e.g.,
Page 11/32
filtering, data compression, data validation, and feature extraction. Most of the time, the required
historical data are unavailable, insufficient, or low quality. The unavailability or low quality of historical
data may not provide the decision-making options to the maintenance engineers especially in preventive
maintenance, where such a data can be used know about the equipment failure, predication of failures in
future and its consequences. However, this historical data is most useful during the predictive analytics to
find the future outcomes[51]. The historical data may also lead to building advanced predictive
maintenance strategies[52] in oil and gas industry.

Similarly, it is not possible to collect data for reliability analysis especially in safety-critical systems due
to catastrophic consequences, where a failure can lead to costing cost human lives such as automotive
systems [53]. The maintenance in the safety critical systems can be maintained by using various
approaches or software programs leveraging reverse engineering[54, 55]. Similarly, risk-based
probabilistic models[56, 57], multi criteria decision making models[58] and reliability based models[59]
can also be applied to provide maintenance tasks in such a environments.

3.4 INDUSTRIAL BLOCKCHAIN-RCM INTEGRATION AND


CHALLENGES
Since, 2014, when Nasdaq announced a partnership with Chain, a Blockchain business, financial
incumbents have been researching Blockchain technology. Since then, the industry has increased its
usage of and investment in the technology, looking into applications in trade finance, commodities
markets, private equity exchanges, digital identity management, and other areas. Blockchain is distributed
database architecture, where information about transactions are stored in digital ledgers [60]. Blockchain
has four major components: distributed shared ledger, permissions, smart contract, and consensus [61]. It
can be applied in Industry 4.0 to reduce the cost, improve productivity and reliability[62]. It can provide
additionally paramount results by ameliorating the tracking hurdles related to the logistical information
of (RCM) functions[63]. In the process of integrating RCM with blockchain, it is imperative to consider the
reliability of “blockchain oracles”. Although, blockchain itself is highly reliable but oracles are the point of
failure and consequently the entire blockchain technology will be failed. In this regard, the reliability-
centered based blockchain models are presented in the literature[64, 65]. Like, RCM, Blockchain also
focuses on higher reliability during the supply chain[66]. Thus, the blend of both technologies will be
providing new opportunities for industry managers to make full use of the available technologies.
Deploying RCM with blockchain can be emerging concept but still there are some challenges during its
deployment with RCM techniques which should be taken into account. These challenges are highlighted
as given below.

A. Data collection issues


In integrating RCM with Industry 4.0, it is inevitable to collect and transmit the core operating status data
and historical failure data over the industrial Internet. These data are core corporate secrets that should
ensure that they are not stolen or tampered with during transmission, so the security of data transmission
is a significant challenge [67]. The blockchain is a promising technique that could tackle this challenge.

Page 12/32
However, there is a lack of mature techniques. Blockchain is a type of distributed accounting technology
jointly maintained by multiple parties, guarantees transmission, and security access through
cryptography. It has the potential to achieve consistent data storage which is difficult to tamper with and
hence prevents denial.

B. Privacy and security challenges


Physical and digital assets vital to industrial enterprises will risk being exposed to the internet. As,
blockchain is distributed architecture so it will be vulnerable to fraudulent attack; and it itself foster
several security challenges[68]. The public blockchain is susceptible to Sybil attack,[69]. However,
blockchain has the potential to provide more secure and robust security frameworks with respect to
maintenance [70–72].

C. Technological and Human perspective


As, blockchain is public technology, which means it provides transparency and redundancy. However, still
there are some government and private companies that are not ready to adopt this technology due to
permission less features and transaction speed[73].

D. Standardization
Standardization efforts are required to make it full-pledged technology and to publish different standards
related to security, interoperability, governance, taxonomy and privacy[74].

E. Storage and data reliability issues


Blockchain is not designed to store bulky data such as produced in the IoT based systems. On the other
hand, the reliability of the data generated by IoT, is among the main challenges in assimilating IoT
technology with the blockchain[75, 76]. Although, blockchain guarantees that data in the chain is
immutable and may identify the transformations, data that has already entered the blockchain remains
corrupt. Apart from harmful situations, there are various reasons for IoT information leakage. It is a
promising technique that could tackle this challenge. For example, a manufacturing process machine
with a unique identity can be registered on a blockchain throughout production, and its performance and
maintenance history can be tracked. When a predictive maintenance alert is issued, a maintenance
service provider could immediately contact via a smart contract, allowing machines to be repaired before
they fail.

Customers might examine the ledger for a product's whole history during the distribution stage,
decreasing counterfeiting and establishing the origin of correctly obtained goods. When a faulty product
is discovered, the producer can search the ledger to rapidly determine the supplier or lousy test results
and notify all recipients of the defective goods.
3.5 INDUSTRIAL ARTIFICIAL INTELLIGENCE-RCM
INTEGRATION AND CHALLENGES

Page 13/32
Predictive maintenance is the potential to use the huge amount of data to forecast and address potential
issues before they cause the failures in operations, services, processes, or systems. Artificial intelligence
and machine learning having robust predictive maintenance technologies which allows businesses to
predict when and where potential service breakdowns is likely to occur and provide quick response to
avoid the interruptions of service. Various new artificial intelligence models have consistently permeated
into industrial applications, giving birth to the emerging cross-integration trend of industrial artificial
intelligence, especially in recent years, with the emerging of new generation of artificial intelligence. By
combining continuous real-time monitoring with predictive maintenance and prescriptive maintenance,
which use AI technologies to recommend addressing incorrect situations, maintenance is improved. With
extremely early detection systems that can differentiate between machine wear and tear and damage-
causing process behavior, it can also lead to the negation of periodic inspection. Manufacturing
organizations understand that unscheduled maintenance and failures reduce profits. These are the key
explanations for why AI/ML platform scores are accessible for testing PdM-related concepts. The
maintenance team can carefully seek to identify the data patterns indicating they intervene to avert
failures using technologies like Google Tensor-Flow, Python, Microsoft Azure, or Amazon Web Services.
An organization may encounter legal concerns due to an AI Application embedded with incorrect data
governance and algorithm. Industrial equipment at the network's edge cannot respond to service requests
in real-time, execute data aggregation, filtering, and prediction, or accomplish complete cloud
coordination without a lightweight computing framework. The ability of industrial AI to analyze huge
amounts of high-dimensional data has the power to disrupt the present maintenance paradigm and take
preventive maintenance systems to new levels.

Following are some of the challenges that are needed to be addressed before implementing AI.

A. Data integration and connectivity


The main challenge is operationalizing predictive maintenance, which entails more than simply
connecting equipment to an AI Platform, streaming data, and analyzing that data [77].

B. Building safety-based AI models


Another challenge is safe AI, since AI is used to predict the maintenance and fault diagnosis, if it is
prediction goes wrong it will cause significant damage to the asset or human. A solution is monitoring
the AI and alarm people if the decision making from AI is violated predefined criteria. For this purpose,
safe AI has been applied to monitor and improve the robustness of safety-critical systems[78]. Here, the
role of AI is to notify, when certain boundary conditions are violated. But, the major limitation of ML-
based monitoring in this study is, it targeted only to the safety evaluation problems. Similarly, another
study presented by Aslansefat et al.[79] Is also utilizing ML-based model for safety-critical systems. The
major challenge related to machine classifier is to take the safe decision, whenever there is sudden shift
in the data. This problem can be resolved by clustering technique[80]. The major challenge related to
dataset characteristics, performance of classification algorithms and features’ selection techniques in
ML-based evaluation model for safety critical systems is an open research question. The potential
solutions are provided by Oreski et al. [81] by developing a knowledge-based system that is focused on
Page 14/32
selecting the best features selection technique among the candidates techniques with respect to new
emerging problem.

C. Expandability issues
While increasing AI Systems' validity might be helpful for several reasons, however, building explainable
AI-based models can be challenging job [82]. In this regard, the financial market practitioners have to put
some efforts to explain the AI-powered models to comprehend their behavior during the normal
conditions of market, in the stress times and managing the associated risks.

D. Data quality
AI and ML models especially rely on clean, accurate and labeled data (in supervised learning) training
data produce more useful results. Access to quality of data is extremely important for the success of any
AI model, but access to such a clean, high quality and labelled data can be challenging[83].

E. Lack of experienced people


The other significant challenge lack of expertise and skillful personnel related to digital transformation
and AI initiatives[84]. It is imperative to have seasoned and professional data scientist, ML experts and
software architects but all of such a personnel is missing and hiring them can be challenging job.

F. Interoperability and technology infrastructure


In manufacturing environment often have heterogeneous tools, machine and production systems that are
utilizing various technologies, some of them may be advanced and some of them may be outdated or
legacy systems, which create compatibility and interoperability issues [84].

3.6 E-Maintenance Challenges


The maintenance procedure is supported by e-Maintenance. E-maintenance represents the technological
services and functions which manage maintenance information. These services can be applied
throughout all the phases of system lifecycle under the umbrella of different maintenance preparation,
execution, assessment and information management[67]. Hence, it has become vital to design an
appropriate e-Maintenance solution by keeping in mind the holistic perspectives and proper
methodologies, strategies, and technologies. During the last few years, e-maintenance is the hot topic in
the industry. However, there are some challenges related to the architecture, infrastructure, content and
contextual; and integration that constrain the motivation of adopting e-maintenance in industry 4.0 that
are discussed below as.

A. Architectural Challenges
Before applying RCM-based technique to conduct e-maintenance in industry 4.0, it is crucial to take into
account the architectural obstacles. Finding problems with a smart maintenance system early on has
been shown to be beneficial. These problems pertain to the design of eMaintenance solutions, such as
creating a complex framework for eMaintenance development, building models that are concentrated on
distributed data processing and analysis, and creating models for the services.

Page 15/32
B. Infrastructural Challenges
According to Service-Oriented Architecture (SOA), the infrastructural challenges may arise whenever the
services are developed and deployed. The Identification and bringing solutions towards the infrastructural
challenges can be more productive to provide solution based on providing the essential tools and
technologies to meet the service requirements [85]. Infrastructure for wired and wireless networks, service
and user authentication, and safety and security mechanisms are some of the primary obstacles. Similar
difficulties might be encountered with the maintainability of eMaintenance services, availability
performance management, and tracing and tracking mechanisms.

C. Content and Contextual


Building RCM based e-maintenance in the industry 4.0 can be challenging job due to the nature of data
collected from the heterogeneous sources. There are various difficulties with this method that must be
determined. These difficulties include developing a suitable ontology that can be utilized to gather
information from integrated data sources such as products, processes, condition monitoring, and
business data. Building a quality assurance system is another way to improve decision-making. Similarly,
there is also lack of establishing a mechanism that can provide a platform to adapt the user’s current
information into the user's context. A sophisticated mechanism is also required to manage uncertainty in
data sets and; a mechanism for describing data in various context and pattern recognition is also a major
concern[86] .

D. Integration Challenges
The eMaintenance also may face some integration challenges, while coordinating, orchestrating and
integrating the data related to maintenance. There are several challenges like service management,
interactivity, configuration management and enabling the potential of integration across a multiplatform
and technologies. Infrastructure for Condition Based Maintenance and Predictive Health Monitoring
approaches must be established in order to improve the decision-making process for eMaintenance. The
merging of data from the business, product, maintenance, condition monitoring, and operation may
effectively accomplish this. It takes comprehensive techniques to create software (S.W.) and hardware
(H.W.) from a holistic system perspective in order to integrate this large data collection and services.
Utilizing a service-oriented architecture is one strategy that may be used in this situation[87]. Data
processing services for eMaintenance are also essential in this context. The fusion of eMaintenance for
the purpose of decision making is as shown in Fig. 2.

The complete summary of all issues identified in this research study is given in Fig. 3. After highlighting
all the major issues involved during the implementation of RCM in Industry 4.0, the next step is to provide
solutions towards the existing challenges and problems.

4 Model-Based Solution
The RCM methods have a close relationship to the sustainability and reliability. An organization that
strives towards the effective sustainability then the effective RCM practices and technologies might be

Page 16/32
considered as they are the most essentials factors. Effective model is required to provide a full-pledged
solution towards the maintenance and reliability issues in the industry 4.0. Several models have been
suggested to provide solution towards the sustainability and reliability of the components of machine in
the industry 4.0. For example, the mathematical models are frequently linked to physical processes that
directly or indirectly impact the condition of equipment components or complete systems in model-based
techniques. Residuals are properties that the analytical model uses to verify the correctness of the
measured results and the anticipated behavior of the process in the context of fault identification and
prognosis. Therefore, the development of methods for precise event detection, fault diagnosis, remedial
solution, and process mining is crucial for the success of novel and automated approaches to the
reliability evaluation of manufacturing systems or any other operating system. Even if exact
mathematical models may not be feasible for many complex activities, such approaches or techniques
are nevertheless relevant if accurate mathematical models can be constructed from physical systems.
Because changes in operational conditions can impact the mathematical model as well. It is impossible
to build a model that can be replicating all the real-world scenarios. Therefore, building such as reliability
based model to address all these factors can be more challenging task.

4.1 Industrial Conceptual Data Model


To address the previously mentioned challenges in Industry 4.0, a conceptual model is presented to
provide reasonable solution towards the existing problems. This model is using all the core technologies
operating in Industry 4.0 such as IoT, AI, DT, Blockchain and Big data for providing intelligent
maintenance. The proposed model also combines the RCM concepts with these.

A. Data collection through IoT


The failure data collection has been the most challenging task in Industry 4.0 for the years but thanks to
IoT to provide all of kind of data such as operational, failure, real time etc. Before, the deployment of IoT
networks, the maintenance teams used to perform manual procedure such as physical inspection of
hardware equipment, but, nowadays, many technologies to provide smart and intelligent maintenance.
Following are the major steps involved in developing the proposed theoretical intelligent predictive
maintenance model. The complete structure of the proposed model is given in Fig. 4.

organizations have adopted IoT-based predictive maintenance in their business strategies to improve the
efficiency and productivity. For example, in oil and Gas Company, IoT has been applied for checking the
working of equipment, fuel level, vibration, leaks etc. PdM relies on the sensor data to optimize the
maintenance activities[88]. Collecting failure data about the equipment is difficult task. In this regard, IoT
offers variety of sensors, actuators, edge devices that are connected to gateway to provide the
operational data to the cloud for various services. In the proposed model of predictive maintenance, IoT
sensors can be used to collect different kinds of data related to the critical equipment’s that can be
further processed by advanced analytics programs to reduce the risk of failure faults. According to our
literature study, several studies have been found that are building models by using IoT for the predictive
maintenance. The complete detail of these studies is given in Table 1.
Page 17/32
Table 1
IoT-based predictive maintenance model
Ref Model description Type of data Application area
collected

[89] IoT-based real time data analytics model Real time and Industrial setup
Historical data

[90] TIP4.0 Industrial IoT predictive maintenance Time series data Industry 4.0
framework

[91] IoT-based predictive maintenance(PdM) Vibration data Healthcare


model for healthcare equipment

[88] Advanced predictive maintenance model Temperature and Manufacturing


using IIoT vibration industry

[92] PdM model for the mechanical equipment Real time Healthcare industry

[93] PdM tool for monitoring equipment using Logistical, testing & Petroleum,
IoT sensing petrochemical & water
treatment

[94] IoT in predictive maintenance for industrial Current data Industrial setup
equipment

[95] IIoT platform for enhancing predictive Historical data Industry 4.0
maintenance

[96] Predictive maintenance based model Temperature and Industry 4.0


leveraging IoT and machine learning vibration data
algorithms

[97] Predictive maintenance using IoT for fleet Proprietary data and Transportation
management routing information

[98] Applying IoT for predictive maintenance of Current data and Coal industry
coal equipment’s historical data

[99] PdM by leveraging IoT big data cloud Historical and real Industrial IoT
system in IIoT time data
B. Data cleaning
The data collected through IoT sensors is to be stored in the database. It is important to store the
important and quality of data related to the equipment in systematic procedure. This procedure is
supported by data cleaning. Data cleaning is process of identifying correct andaccurate data by removing
inappropriate and incomplete, and inaccurate data [100]. Data cleaning allows storing data in quality and
structured form in the database. During this procedure, data is converted into standard form such as
elimination of redundancies; removal of unnecessary, handling missing data.

It can be achieved by using machine learning technologies or big data or both technologies combined
together. There are various studies available in literature that is providing intelligent method for pre-

Page 18/32
processing of data. For example, the work conducted by Bekar et al.[101] is presenting the smart
approach for cleaning data using machine learning approach. Gudivada et al. [102] presents study that
combines both Big data and ML techniques for cleaning data. Big data can also be applied as an ideal
choice for cleaning data[103].

C. Storing data using block chain


The data collected from IoT sensors is required to be securely stored in the database. This can be
achieved by using Blockchain chain technology.

The major objective of applying Blockchain is not only to provide decentralized secure storage but it also
provides massive support to DT, with enormous benefits such as transparency, DTs data traceability, DTs
access privileges, securing and trustworthy DT data, decentralized infrastructure and automation[104–
106]. Blockchain technology enables DTs to draw the most important conclusions such as identifying
faults and recommending precautionary measures[107]. Similarly, Blockchain-based DTs allows
continuous data acquisition in industry 4.0[106].

D. Intelligent predictive maintenance using RCM


This is the integral phase of the proposed model, in which intelligence maintenance strategy is prepared
by integrating RCM and Industry 4.0 technologies. The major reason for applying RCM is, it provides a
solid foundation for building intelligent and smart maintenance strategy. RCM is applied to collect detail
about the functions of equipment, where a decision-making will be performed about the health of
equipment. The analysis and evaluation of equipment is carried out by RCM approach supported by
industry 4.0. The functions of equipment are identified by using FMEA. The basic input is provided to the
proposed maintenance model is description of equipment and current data. RCM helps in investigating
the causes and effects of failure in equipment. This part of the proposed model consists of the following
steps.

E. Failure diagnosis
Failure diagnosis is the most important and initial step of predictive maintenance. In the proposed
conceptual data model, failure diagnosis will be performed by using FMEA supported by artificial
intelligence technology. The efficiency and accuracy of RCM can be improved by using artificial
intelligence. AI has already been considered as best technological solution to enhance the FMEA of RCM
process [16]. The combination of FMEA and AI has significant outcomes towards risk management and
reliability[108]. Therefore, various failure modes in the proposed model will be identified and then ranking
will be done by using grey relation analysis (GRA) to understand the level of risks associated with faults.
The machine learning approach will be applied to extract the features. The application of AI in fault
diagnosis yielded promising results and there are several studies available in literature[109, 110].

F. Failure prediction
It is important to monitor the equipment or asset for any kind of failure before they occur. The application
of AI in the proposed model will help to develop an intelligent predictive model. In the proposed model the
historical data will be provided as input. The AI model will take input data by making prediction based on
Page 19/32
hypothesis inferred from the data. The extraction of more relevant data to be provided to the proposed
model can be challenging. It requires more time and energy to preprocess the data before it is provided as
input to get the most desirable results about the failure. This problem can be tackled by using RCM that
helps in analyzing the equipment failure. Similarly, the proposed model has the ability to apply Big data
and ML techniques to maintain the quality of data. There are various machine learning approaches that
can be proven as effective such as support vector machine (SVM), artificial neural network (ANNs), K-
nearest neighbors(KNN) and decision trees [111, 112]. Among the machile learning models, SVM has
demonstrated significant performance comparatively. This approach is more ideal for time-series
data[111]. Similarly, its simple nature and functionality with high dimension data make it also a helpful
option[112].

G. Failure prognosis

This is the final stage of the proposed model, where the proposed model will predict failure as it occurs.
The proposed model will apply machine learning approaches to predict the failure of the equipment. RCM
will also be applied to evaluate and analyze the failure risk and cost. RCM-based maintenance strategy
will be used to optimize the maintenance program.

5 Lesson Learned
As, Industry 4.0 is adopting various cutting edge technologies such as IoT, Big data, Blockchain, Artificial
Integlligence, Digital Twin etc. so the integration of RCM for the smart maintenance with these individual
technologies will create many challenges for the manangers and stakeholders. Therefore, it requires a
sound knowledge to identify and highlight the challenges and take full advatange of RCM, when it is
applied in the background of Indsustry 4.0. This review paper is more focused on identifying the
underlying challenges related to maintenance by applying RCM in industry 4.0 and providing tentative
solutions to overcome these challenges. These challenges vary in nature, when RCM is applied with every
different technology for the purpose of efficient maintenance related tasks. For example, integrating RCM
with IoT will require to address challenges related to security, building a reference architecture and
standardization, fault predication and limited applicability. Similarly, challenges that may arise during the
amalgam of RCM and digital twins are relalted to complex modeling and data flow, data analysis security
and technology adoption, culture and lack of skilled personnel. Challenges which need to be overcome,
while integrating RCM with Block chain technology are about analyzing the collected data, privacy and
security challenges, technological and human perspective and standardization; and storage and
reliability. Implementing RCM with AI/ML approaches also have some concerns related to data
integration and connectivity, designing safety-based AI models, expandability issues, data quality, lack of
experienced people and interoperability and technology infrastructure. This study also identifies
challenges faced by RCM integrated for e-maintenance, these challenges include architectural challenges,
infrastructural challenges, content and contextual and integration challenges.

Page 20/32
This work provides solid foundations for the researchers and industry managers to apply the concepts
and methods of RCM with respect to maintenance in the industry 4.0. This work is also helpul in
providing solutions towards the bottleneck of RCM applications. This review identifies the most emerging
challenges in comparison to the previous studies and it puts forward an innovative concept by presenting
a conceputal model to overcome the emerging challenges related to maintenance. Smart and predictive
maintenance can be achieved by combining RCM with Industry 4.0 set up.

6 Conclusion
The implementation of RCM in industry 4.0 for the maintenance can be more effective to provide most
efficient maintenance strategies and optimized maintenance through predictive fault identification of
asset or component with the factor of reliability. However, its implementation also brings many
challenges that are required to be highlighted. The major theme of this article is to investigate and
highlight the challenges that can be emerging while collaborating RCM with the technologies such as IoT,
DTs, Blockchain, Big data, and Artificial Intelligence etc. operating in the Industry 4.0. This paper also
presents a model-based solution to provide solution towards the existing challenges that may arise due
to integration of RCM in the background of Industry 4.0. Therefore, a conceptual model is put forward
that integrates RCM process in the industry 4.0 to overcome these challenges.

The future focus of our study is to implement the reliability-centered based maintenance approaches in
industry 4.0. We also focus on validating the proposed model that can be applied in practical scenarios in
industry 4.0 or above.

Declarations
Ethical Approval

Dear Editor-in-Chief,

I am writing to seek ethical approval for the submission of a research article entitled "Implementation of
Reliability Centered Maintenance (RCM) in the background of industry 4.0 –Issues, Challenges, and
Opportunities" to be considered for publication in The Journal of Supercomputing. The research
described in the manuscript adheres to the highest ethical standards and aligns with the guidelines
esteemed journal. We would be happy to provide any additional information or documentation to support
the ethical integrity of our research if required.

Competing Interests

The authors declare that they have no known competing financial interests or personal relationships that
could have appeared to influence the work reported in this paper.

Author contribution

Page 21/32
Khalil Ur Rahman: The first author of an academic paper typically makes significant contributions in
conceptualization, methodology, and writing - original draft preparation. He developed the initial research
idea or concept, contributed to the study design, participant recruitment, data collection, and analysis, and
drafted the manuscript's various sections, including the introduction, methods, results, and discussion. He
also collaborated with other authors to ensure the research was conducted ethically and rigorously and
that the manuscript accurately reflected the research findings.

Professor Mian Zhibao*: The corresponding author of an academic paper typically makes significant
contributions to supervision, investigation, and reviewing. He has provided oversight and guidance
throughout the research project, ensuring the study was conducted rigorously and ethically. He also
contributed to the investigation, providing resources or securing funding for the research project and
contributing to interpreting the research findings. Additionally, the corresponding author is often
responsible for reviewing the manuscript, providing feedback to the other authors, and working with the
journal editor to address any reviewer comments or concerns. Their contributions are critical in ensuring
the research is conducted and reported thoroughly and rigorously.

Yasir Ali: The third author of an academic paper has contributed significantly to ensuring the accuracy
and clarity of the research work, as well as providing critical feedback to the other first authors.
Additionally, He has contributed to reviewing and editing the manuscript, working with the fisrt authors to
ensure that it accurately reflects the research findings and meets the intended journal or publication
standards.

Funding Information

This work was supported in part by the Industry Support Plan Project of Gansu Province (grant
#2021CYZC-06), the NSFC (grant #12061065), Ningxia Key Research and Development Plan Project
(grant #2018BEE03019), the Young Doctor Research Funding grant of Gansu Province (grant #2021QB-
022), and the Science and Technology Program of Gansu Province (grant #21JR7RA135).

Availability of data and materials

Not Applicable.

References
1. Geisbush, J., Performing a Reliability Centered Maintenance Workshop on the World’s Largest
Prestressed Concrete Pipes, in Pipelines 2020. 2020, American Society of Civil Engineers Reston, VA.
p. 343-351.
2. Sajaradj, Z., L.N. Huda, and S. Sinulingga. The Application of Reliability Centered Maintenance
(RCM) Methods to Design Maintenance System in Manufacturing (Journal Review). in IOP
Conference Series: Materials Science and Engineering. 2019. IOP Publishing.

Page 22/32
3. Farsi, M., R.K. Mishra, and J.A. Erkoyuncu, Industry 5.0 for sustainable reliability centered
maintenance. Available at SSRN 3944533, 2021.
4. Sarjono, H., M.M. Trisnianto, and B.L. Handoko. Machine Maintenance Planning with Reliability
Centered Maintenance to Increase Product Quality Reliability in The Indonesian Automotive Industry
Systematic Literature Review. in 2021 3rd International Conference on E-Business and E-commerce
Engineering. 2021.
5. Alrifaey, M., et al., Optimization and selection of maintenance policies in an electrical gas turbine
generator based on the hybrid reliability-centered maintenance (RCM) model. Processes, 2020. 8(6):
p. 670.
6. Catelani, M., et al., Optimizing maintenance policies for a yaw system using reliability-centered
maintenance and data-driven condition monitoring. IEEE Transactions on Instrumentation and
Measurement, 2020. 69(9): p. 6241-6249.
7. Márquez, F.P.G.a., F. Schmid, and J.C. Collado, A reliability centered approach to remote condition
monitoring. A railway points case study. Reliability Engineering & System Safety, 2003. 80(1): p. 33-
40.
8. Dialynas, E.N., L.G. Daoutis, and T.D. Diagoupis, Implementation of reliability centered maintenance
programs into electric power systems.
9. Huang, L., et al. Discussion on application of reliability-centered maintenance to reliability
improvement in new nuclear power plants. in 2013 International Conference on Quality, Reliability,
Risk, Maintenance, and Safety Engineering (QR2MSE). 2013. IEEE.
10. Fischer, K., F. Besnard, and L. Bertling, Reliability-centered maintenance for wind turbines based on
statistical analysis and practical experience. IEEE Transactions on Energy Conversion, 2011. 27(1): p.
184-195.
11. Teixeira, H.N., I. Lopes, and A.C. Braga, Condition-based maintenance implementation: a literature
review. Procedia Manufacturing, 2020. 51: p. 228-235.
12. Close, T., Ask an expert: What Industry 4.0 can do for maintenance. McKinsey & Company, 2017.
13. Frank, A.G., L.S. Dalenogare, and N.F. Ayala, Industry 4.0 technologies: Implementation patterns in
manufacturing companies. International Journal of Production Economics, 2019. 210: p. 15-26.
14. Das, S. and S. Kumar, Applying RCM Techniques to Household Consumer Durables through the
Internet of Things. International Journal of Mechanical and Production Engineering Research and
Development (IJMPERD), 2013. 3(3): p. 117-130.
15. Park, S.C., Y.J. Kim, and J.U. Won, Application of IoT for the maintaining rolling stocks. Quality
Innovation Prosperity, 2017. 21(2): p. 71-83.
16. Venkataswamy, R., K.U. Rao, and P. Meena. Internet of things based metaheuristic reliability centered
maintenance of distribution transformers. in IOP Conference Series: Earth and Environmental
Science. 2020. IOP Publishing.
17. Rahman, A., et al. Industry 4.0 and society 5.0 through lens of condition based maintenance (CBM)
and machine learning of artificial intelligence (MLAI). in IOP Conference Series: Materials Science
Page 23/32
and Engineering. 2020. IOP Publishing.
18. Jafarpisheh, R., M. Karbasian, and M. Asadpour, A hybrid reliability-centered maintenance approach
for mining transportation machines: a real case in Esfahan. International Journal of Quality &
Reliability Management, 2020.
19. Compare, M., P. Baraldi, and E. Zio, Challenges to IoT-enabled predictive maintenance for industry 4.0.
IEEE Internet of Things Journal, 2019. 7(5): p. 4585-4597.
20. Achouch, M., et al., On Predictive Maintenance in Industry 4.0: Overview, Models, and Challenges.
Applied Sciences, 2022. 12(16): p. 8081.
21. Hussain, Z. and H. Jan, Establishing simulation model for optimizing efficiency of CNC machine
using reliability-centered maintenance approach. International Journal of Modeling, Simulation, and
Scientific Computing, 2019. 10(06): p. 1950034.
22. Mubarak, A., et al. Digital Twin Enabled Industry 4.0 Predictive Maintenance Under Reliability-Centred
Strategy. in 2022 First International Conference on Electrical, Electronics, Information and
Communication Technologies (ICEEICT). 2022. IEEE
23. Liu, Z., N. Meyendorf, and N. Mrad. The role of data fusion in predictive maintenance using digital
twin. in AIP conference proceedings. 2018. AIP Publishing LLC.
24. Cheng, Z. and X. Jia. An intelligent reliability centered maintenance analysis system based on case-
based reasoning & rule-based reasoning. in Annual Reliability and Maintainability Symposium, 2005.
Proceedings. IEEE.
25. Cheng, Z., et al., A framework for intelligent reliability centered maintenance analysis. Reliability
engineering & system safety, 2008. 93(6): p. 806-814.
26. Li, H., et al., Improving rail network velocity: A machine learning approach to predictive maintenance.
Transportation Research Part C: Emerging Technologies, 2014. 45: p. 17-26.
27. Saidi, R. and A. Soulhi, Why the approach big data, is essential for the reliability-centered
maintenance. International journal of engineering science & research technology, 2017.
28. Jiang, Y., et al., A review on soft sensors for monitoring, control, and optimization of industrial
processes. IEEE Sensors Journal, 2020. 21(11): p. 12868-12881.
29. Ghadge, A., et al., Link between Industry 4.0 and green supply chain management: Evidence from the
automotive industry. Computers & Industrial Engineering, 2022: p. 108303.
30. Chris Coleman, J.B., Using prescriptive maintenance to improve asset performance. 2021, TATA
Consultancy Services.
31. Girma, A., Analysis of Security Vulnerability and Analytics of Internet of Things (IOT) Platform, in
Information Technology-New Generations. 2018, Springer. p. 101-104.
32. Reyes-Picknell, J., IIoT for Predictive Maintenance and Big Data. 2020, Accendo Reiliability.
33. Mobley, R.K., An introduction to predictive maintenance. 2002: Elsevier.
34. Bertino, E. Security and Privacy in the IoT. in International Conference on Information Security and
Cryptology. 2017. Springer.

Page 24/32
35. Tekeoglu, A. and A.Ş. Tosun. An Experimental Framework for Investigating Security and Privacy of
IoT Devices. in International Conference on Intelligent, Secure, and Dependable Systems in Distributed
and Cloud Environments. 2017. Springer.
36. Gurunath, R., et al. An Overview: Security Issue in IoT Network. in 2018 2nd International Conference
on I-SMAC (IoT in Social, Mobile, Analytics and Cloud)(I-SMAC) I-SMAC (IoT in Social, Mobile,
Analytics and Cloud)(I-SMAC), 2018 2nd International Conference on. 2018. IEEE.
37. Weyrich, M. and C. Ebert, Reference architectures for the internet of things. IEEE Software, 2015.
33(1): p. 112-116.
38. Li, J., et al., Towards decision support systems for energy management in the smart industry and
Internet of Things. Computers & Industrial Engineering, 2021. 161: p. 107671
39. van Dinter, R., B. Tekinerdogan, and C. Catal, Predictive maintenance using digital twins: A systematic
literature review. Information and Software Technology, 2022: p. 107008.
40. Aivaliotis, P., K. Georgoulias, and G. Chryssolouris, The use of Digital Twin for predictive maintenance
in manufacturing. International Journal of Computer Integrated Manufacturing, 2019. 32(11): p.
1067-1080.
41. Mihai, S., et al., A digital twin framework for predictive maintenance in industry 4.0.
42. Scheidegger, A.P.G., et al., An introductory guide for hybrid simulation modelers on the primary
simulation methods in industrial engineering identified through a systematic review of the literature.
Computers & Industrial Engineering, 2018. 124: p. 474-492.
43. de Sousa Junior, W.T., et al., Discrete simulation-based optimization methods for industrial
engineering problems: A systematic literature review. Computers & Industrial Engineering, 2019. 128:
p. 526-540.
44. Hartmann, D. and H. Van der Auweraer, Digital twins, in Progress in Industrial Mathematics: Success
Stories. 2021, Springer. p. 3-17.
45. Sharma, A., et al., Digital twins: State of the art theory and practice, challenges, and open research
questions. Journal of Industrial Information Integration, 2022: p. 100383.
46. Segovia, M. and J. Garcia-Alfaro, Design, Modeling and Implementation of Digital Twins. Sensors,
2022. 22(14): p. 5396.
47. Wahid, A. and K. Kashyap, Cassandra—A distributed database system: An overview. Emerging
technologies in data mining and information security, 2019: p. 519-526.
48. Qi, Q., et al., Enabling technologies and tools for digital twin. Journal of Manufacturing Systems,
2021. 58: p. 3-21.
49. Wagner, R., et al., Challenges and potentials of digital twins and industry 4.0 in product design and
production for high performance products. Procedia CIRP, 2019. 84: p. 88-93.
50. Durão, L.F., et al. Digital twin requirements in the context of industry 4.0. in IFIP international
conference on product lifecycle management. 2018. Springer.

Page 25/32
51. Jiang, F., et al., Digital twin and its implementations in the civil engineering sector. Automation in
Construction, 2021. 130: p. 103838.
52. Broo, D.G. and J. Schooling, Digital twins in infrastructure: definitions, current practices, challenges
and strategies. International Journal of Construction Management, 2021: p. 1-10.
53. Oliveira-Dias, D., J.M. Maqueira, and J. Moyano-Fuentes, The link between information and digital
technologies of industry 4.0 and agile supply chain: Mapping current research and establishing new
research avenues. Computers & Industrial Engineering, 2022: p. 108000.
54. Lee, C., Y. Cao, and K.H. Ng, Big data analytics for predictive maintenance strategies, in Supply Chain
Management in the Big Data Era. 2017, IGI Global. p. 50-74.
55. Sajid, S., et al., Data science applications for predictive maintenance and materials science in context
to Industry 4.0. Materials today: proceedings, 2021. 45: p. 4898-4905.
56. Mohr, M., et al., Towards collaborative predictive maintenance leveraging private cross-company
data. INFORMATIK 2020, 2021.
57. Kumar, U. and D. Galar, Maintenance in the era of industry 4.0: issues and challenges. Quality, IT and
business operations, 2018: p. 231-250.
58. Wu, S.-j., et al., A neural network integrated decision support system for condition-based optimal
predictive maintenance policy. IEEE Transactions on Systems, Man, and Cybernetics-Part A: Systems
and Humans, 2007. 37(2): p. 226-236.
59. Reyna, A., et al., On blockchain and its integration with IoT. Challenges and opportunities. Future
generation computer systems, 2018. 88: p. 173-190.
60. Karim, R., et al., Maintenance analytics–the new know in maintenance. IFAC-PapersOnLine, 2016.
49(28): p. 214-219.
61. Amruthnath, N. and T. Gupta. A research study on unsupervised machine learning algorithms for
early fault detection in predictive maintenance. in 2018 5th international conference on industrial
engineering and applications (ICIEA). 2018. IEEE.
62. Giunta, G., et al. Digital transformation of historical data for advanced predictive maintenance. in
Abu Dhabi International Petroleum Exhibition & Conference. 2020. OnePetro.
63. Lazarova-Molnar, S. and N. Mohamed, Reliability assessment in the context of industry 4.0: data as a
game changer. Procedia Computer Science, 2019. 151: p. 691-698.
64. Gannod, G.C. and B.H. Cheng, Facilitating the Maintenance of Safety-Critical Systems. International
Journal of Software Engineering and Knowledge Engineering, 1994. 4(02): p. 183-204.
65. Bennett, K. and M. Ward, Using formal transformations for the reverse engineering of real-time safety
critical systems, in Technology and Assessment of Safety-Critical Systems. 1994, Springer. p. 204-
223.
66. Xu, T., et al., Risk-based predictive maintenance for safety-critical systems by using probabilistic
inference. Mathematical Problems in Engineering, 2013. 2013.

Page 26/32
67. Sanne, J.M., Framing risks in a safety‐critical and hazardous job: risk‐taking as responsibility in
railway maintenance. Journal of Risk Research, 2008. 11(5): p. 645-658.
68. Arjomandi, M.A., et al., A fuzzy DEMATEL-ANP-VIKOR analytical model for maintenance strategy
selection of safety critical assets. Advances in Mechanical Engineering, 2021. 13(4): p.
1687814021994965.
69. Jia, X. and L. Cui, Optimization of joint maintenance strategy for safety‐critical systems with
different reliability Expert Systems, 2011. 28(3): p. 199-208.
70. Monrat, A.A., O. Schelén, and K. Andersson, A survey of blockchain from the perspectives of
applications, challenges, and opportunities. IEEE Access, 2019. 7: p. 117134-117151.
71. Agrawal, T.K., et al., Blockchain-based framework for supply chain traceability: A case example of
textile and clothing industry. Computers & industrial engineering, 2021. 154: p. 107130.
72. Mohamed, N. and J. Al-Jaroodi. Applying blockchain in industry 4.0 applications. in 2019 IEEE 9th
annual computing and communication workshop and conference (CCWC). 2019. IEEE.
73. Mudliar, K., H. Parekh, and P. Bhavathankar. A comprehensive integration of national identity with
blockchain technology. in 2018 International Conference on Communication information and
Computing Technology (ICCICT). 2018. IEEE.
74. Lo, S.K., et al., Reliability analysis for blockchain oracles. Computers & Electrical Engineering, 2020.
83: p. 106582.
75. Ravaghi Ardabili, H.A., M.R. Haghifam, and S.M. Abedi, A probabilistic reliability‐centred maintenance
approach for electrical distribution networks. IET Generation, Transmission & Distribution, 2021.
15(6): p. 1070-1080.
76. Korepin, V., et al., Improving supply chain reliability with blockchain technology. Maritime Economics
& Logistics, 2021: p. 1-14.
77. Ni, J., X. Gu, and X. Jin, Preventive maintenance opportunities for large production systems. CIRP
Annals, 2015. 64(1): p. 447-450.
78. Cañas, H., et al., Implementing Industry 4.0 principles. Computers & industrial engineering, 2021. 158:
p. 107379.
79. Halpin, H. and M. Piekarska. Introduction to Security and Privacy on the Blockchain. in 2017 IEEE
European Symposium on Security and Privacy Workshops (EuroS&PW). 2017. IEEE. Aleshi, A., R.
Seker, and R.F. Babiceanu. Blockchain model for enhancing aircraft maintenance records security. in
2019 IEEE International Symposium on Technologies for Homeland Security (HST). 2019. IEEE.

81. Agarwal, V. and S. Pal. Blockchain meets IoT: a scalable architecture for security and maintenance.
in 2020 IEEE 17th International Conference on Mobile Ad Hoc and Sensor Systems (MASS). 2020.
IEEE.
82. Lee, J., M. Azamfar, and J. Singh, A blockchain enabled Cyber-Physical System architecture for
Industry 4.0 manufacturing systems. Manufacturing letters, 2019. 20: p. 34-39.

Page 27/32
83. Akram, S.V., et al., Adoption of blockchain technology in various realms: Opportunities and
challenges. Security and Privacy, 2020. 3(5): p. e109.
84. Deepa, N., et al., A survey on blockchain for big data: approaches, opportunities, and future
directions. Future Generation Computer Systems, 2022.
85. Sial, M.F.K., Blockchain Technology–Prospects, Challenges and Opportunities. IEEE Blockchain,
Technical Briefs, 2019.
86. Uddin, M.A., et al., A survey on the adoption of blockchain in iot: Challenges and solutions.
Blockchain: Research and Applications, 2021. 2(2): p. 100006.
87. Haddud, A., et al., Examining potential benefits and challenges associated with the Internet of Things
integration in supply chains. Journal of Manufacturing Technology Management, 2017.
88. Aslansefat, K., et al. SafeML: safety monitoring of machine learning classifiers through statistical
difference measures. in International Symposium on Model-Based Safety and Assessment. 2020.
Springer.
89. Aslansefat, K., et al., Toward improving confidence in autonomous vehicle software: A study on
traffic sign recognition systems. Computer, 2021. 54(8): p. 66-76.
90. Abdullatif, A., F. Masulli, and S. Rovetta, Clustering of nonstationary data streams: A survey of fuzzy
partitional methods. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 2018.
8(4): p. e1258.
91. Oreski, D., S. Oreski, and B. Klicek, Effects of dataset characteristics on the performance of feature
selection techniques. Applied Soft Computing, 2017. 52: p. 109-119.
92. Nassr, I.K., Artificial Intelligence, Machine Learning and Big Data I Finance Opportunities, Challenges
and Implications for Policy Makers. 2021: OECD.
93. Peres, R.S., et al., Industrial artificial intelligence in industry 4.0-systematic review, challenges and
outlook. IEEE Access, 2020. 8: p. 220121-220139.
94. Fujimaki, R., The 6 Challenges of Implementing AI in Manufacturing. 2020, American Machinist.
95. Lam, W.-o., Reliability centred maintenance of precision rail track point machine.
96. Azhar, M., M. Omar, and A. Shaiful. A case study of industry 4.0 adoption for the manufacturing
SMEs in Malaysia–Are we ready? in AIP Conference Proceedings. 2021. AIP Publishing LLC.
97. Civerchia, F., et al., Industrial Internet of Things monitoring solution for advanced predictive
maintenance applications. Journal of Industrial Information Integration, 2017. 7: p. 4-12.
98. Civerchia, F., et al., Industrial Internet of Things monitoring solution for advanced predictive
maintenance applications. Journal of Industrial Information Integration, 2017. 7: p. 4-12.
99. Mihigo, I.N., et al., On-Device IoT-Based Predictive Maintenance Analytics Model: Comparing
TinyLSTM and TinyModel from Edge Impulse. Sensors, 2022. 22(14): p. 5174.
100. Shamayleh, A., M. Awad, and J. Farhat, IoT based predictive maintenance management of medical
equipment. Journal of medical systems, 2020. 44(4): p. 1-12.

Page 28/32
101. Niyonambaza, I., M. Zennaro, and A. Uwitonze, Predictive Maintenance (PdM) Structure Using
Internet of Things (IoT) for Mechanical Equipment Used into Hospitals in Rwanda. Future Internet,
2020. 12(12): p. 224.
102. Bayoumi, A. and R. McCaslin, Internet of things–a predictive maintenance tool for general machinery,
petrochemicals and water treatment, in Advanced Technologies for Sustainable Systems. 2017,
Springer. p. 137-146.
103. Parpala, R.C. and R. Iacob. Application of IoT concept on predictive maintenance of industrial
equipment. in MATEC Web of Conferences. 2017. EDP Sciences.
104. Chehri, A. and G. Jeon, The industrial internet of things: examining how the IIoT will improve the
predictive maintenance, in Innovation in medicine and healthcare systems, and multimedia. 2019,
Springer. p. 517-527.
105. Samatas, G.G., S.S. Moumgiakmas, and G.A. Papakostas. Predictive Maintenance-Bridging Artificial
Intelligence and IoT. in 2021 IEEE World AI IoT Congress (AIIoT). 2021. IEEE.
106. Killeen, P., et al., IoT-based predictive maintenance for fleet management. Procedia Computer Science,
2019. 151: p. 607-613.
107. Dong, L., R. Mingyue, and M. Guoying, Application of internet of things technology on predictive
maintenance system of coal equipment. Procedia engineering, 2017. 174: p. 885-889.
108. Truong, H.-L. Integrated analytics for IIoT predictive maintenance using IoT big data cloud systems.
in 2018 IEEE International Conference on Industrial Internet (ICII). 2018. IEEE.
109. Traini, E., et al., Machine learning framework for predictive maintenance in milling. IFAC-
PapersOnLine, 2019. 52(13): p. 177-182.
110. Bekar, E.T., P. Nyqvist, and A. Skoogh, An intelligent approach for data pre-processing and analysis in
predictive maintenance with an industrial case study. Advances in Mechanical Engineering, 2020.
12(5): p. 1687814020919207.
111. Gudivada, V., A. Apon, and J. Ding, Data quality considerations for big data and machine learning:
Going beyond data cleaning and transformations. International Journal on Advances in Software,
2017. 10(1): p. 1-20.
112. Xu, X., Y. Lei, and Z. Li, An incorrect data detection method for big data cleaning of machinery
condition monitoring. IEEE Transactions on Industrial Electronics, 2019. 67(3): p. 2326-2336.
113. Suhail, S., et al., Blockchain-based digital twins: research trends, issues, and future challenges. ACM
Computing Surveys (CSUR), 2021.
114. Yaqoob, I., et al., Blockchain for digital twins: Recent advances and future research challenges. IEEE
Network, 2020. 34(5): p. 290-298.
115. Sahal, R., et al., Blockchain-empowered digital twins collaboration: smart transportation use case.
Machines, 2021. 9(9): p. 193.
116. Suhail, S., et al., Trustworthy digital twins in the industrial internet of things with blockchain. IEEE
Internet Computing, 2021.

Page 29/32
117. Duan, C., et al., Reliability Analysis of Intelligent Manufacturing Systems Based on Improved FMEA
Combined With Machine Learning.
118. Lei, Y., et al., Applications of machine learning to machine fault diagnosis: A review and roadmap.
Mechanical Systems and Signal Processing, 2020. 138: p. 106587.
119. Liu, R., et al., Artificial intelligence for fault diagnosis of rotating machinery: A review. Mechanical
Systems and Signal Processing, 2018. 108: p. 33-47.
120. das Chagas Moura, M., et al., Failure and reliability prediction by support vector machines regression
of time series data. Reliability Engineering & System Safety, 2011. 96(11): p. 1527-1534.
121. Fox, H., et al., A review of predictive and prescriptive offshore wind farm operation and maintenance.
Energies, 2022. 15(2): p. 504.

Figures

Figure 1

Predictive maintenance process

Page 30/32
Figure 2

Fusion of eMaintenance for decision support

Figure 3

Summary of challenges identified

Page 31/32
Figure 4

Proposed model for predictive maintenance using RCM and Industry 4.0

Page 32/32

You might also like